extract_profiles function returning wrong connections when neighbouring cells drain to outlet

[barbedorafael]

I'm trying to get upstream/downstream relationships of several river stretches spread across several catchments in a large region. However, I noticed a problem when the most downstream cell of two neighboring catchments drain to the same outlet point, which happens quite often: instead of considering both cells as the end of the domain (outlet point), one catchment is considered as a contributor to the other, which is not true, and oftentimes the larger catchment is the contributor to the smaller one, which has no topological sense.

Below I attached an image ilustrating the issue, where the thicker line is wrongly considered to be downstream of the larger catchment.

I couldn't solve that in the code, so right now I'm just manually adjusting the values to fit my needs. I'm reporting this hoping it can be solved in the future. Thanks.